Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Part III Supports for Working Families Part III reviews the public supports currently available in the United States to families with working parents. Chapter 8 considers the public policies, including leave policies, tax policies, and education programs, available to working families and the implications of these policies for child and adolescent well-being. Chapter 9 summarizes the committee's findings and conclusions, as well as some policy options that are warranted, in the committee's view, by these trends and research evidence.
8 Policies to Support Working Families T his chapter reviews existing support policies available in the United States to families with working parents. The primary focus is on public policies--specifically leave policies, tax poli- cies, education programs, and programs to assist families in paying for child care. For each major policy area, the support provided and the families who benefit from it are described. We conclude by briefly consid- ering the implications of the current patterns of support for child and adolescent development and public policy. There is some evidence that family support policies have also been integrated into employment policies of private-sector companies (Galinsky et al., 1992; Galinsky and Friedmand, 1993). Only limited data are available on these employer policies and how well they meet the needs of children in working families. The data that do exist suggest that access to corporate policies and benefits is uneven, with lower-income workers less likely to have coverage (see Table 8-1). How- ever, overall, the data are limited and do not provide a comprehensive understanding of who these policies affect and the extent to which they support the well-being of children in working families. The material pre- sented here and the committee's findings and conclusions are therefore focused on public policies. LEAVE POLICIES Leave policies give working parents the right to take time off from work without the risk of losing their jobs. Evidence presented in Chapter 4 229
230 WORKING FAMILIES AND GROWING KIDS Six /A Daily Flextime (%) 26 20 31 N/A N/A N N/A 18 children 19 44 13 22 47 Age with Under fathers older. /A Traditional Flextime (%) 44 39 48 N/A N/A N N/A 42 and and 35 61 31 41 62 18 Children mothers ages Leave Sick Policy with Paid for Children (%) 49 N/A N/A 34 51 37 51 37 48 61 36 48 66 the women to employed and Access men Employees Paid Holidays (%) 84 80 88 63 87 71 86 67 includes 87 93 74 85 93 by with sample employed Policies Employees Paid Vacation Days (%) 85 78 89 57 89 58 89 69 This 88 91 78 86 88 3,552 of of values. Place sample Percentage Family Health Insurance (%) Work 86 78 89 57 89 73 86 66 87 95 69 86 93 specific on national data * 69-72 77-79 Corporate Sample Size 513-536 228-231 303-306 450-462 443-456 115-122 247-254 124-126 109-116 280-293 106-120 selected (2000). missing to to Bond due randomly a available. and Access children year vary not $71,500 with $19.25 per from 8-1 Group six to sizes six Galinsky Status Earnings to age Status Income age indicates parents Mothers Fathers Part-time Full-time Single Married/partnered $7.70 $7.71 $19.25 <$28,000 $28,000 $71,600 *Sample N/A TABLE Employee All under Gender Work Marital Hourly Family under SOURCE:
POLICIES TO SUPPORT WORKING FAMILIES 231 shows that very young children may be particularly affected by maternal employment, and, for newborns, outcomes for mothers and children are better when mothers are able to take more than 12 weeks of leave. Out- comes for children may be better when mothers are able to return to work part time or to delay returning to work full time until after the first year. There are many types of leave--vacation leave, personal leave, sick leave, leave for jury service, leave for bereavement, and so on. Here we consider family leave and medical leave--the major types of leave that working parents may need to take to care for their children and adolescents. Family Leave Family leave includes several types of leave that families use to care for children and adolescents, including most commonly maternity leave but also paternity leave, leave to care for a sick child or adolescent, or leave to arrange care for a child. In the area of family leave, as with other types of leave, the United States historically has not had many public policies; rather, leave policies have mainly been left to the discretion of employers. The United States had no national maternity leave legislation until the passage of the Family and Medical Leave Act (FMLA) in 1993. Prior to 1993, the United States did have some state family leave laws, which provided the right to a job-protected leave for maternity and pater- nity to some mothers and fathers in some states. These state laws vary by their effective date, the type and size of firms covered, the number of weeks of leave provided, and the job tenure and working hours requirements that employees must satisfy in order to be eligible for coverage (Han and Waldfogel, 2002). There is also a handful of states that have temporary disability insur- ance laws providing the right to a paid temporary leave for disability, including disability associated with maternity. In these states, employers are reimbursed for a share of the costs of providing a paid leave for mater- nity during the period of medically certified disability (usually 6 weeks, 8 weeks if delivery was by Caesarean section). Since 1978, with the passage of the Pregnancy Disability Act, the federal government has mandated that disability programs such as these must cover maternity like any other form of disability, but the act does not require firms or states to have disability programs in the first place. Prior to the passage of the FMLA, as a result of the limited number and scope of state laws, many employed women had no right to job-protected maternity leave, and coverage levels were particularly low among part-time employees and those working for small firms (see Table 8-2). The share of men with paternity leave coverage was even lower, and it was lowest among those working part time or in small firms.
232 WORKING FAMILIES AND GROWING KIDS TABLE 8-2 Percentage of Private-Sector Employees with Family Leave Coverage, 1991 to 1997 1991 1993 1995 1997 A. Full-time employees, medium-sized and large establishments Maternity leave coverage: Percentage with unpaid leave 37 60 84 93 Percentage with paid leave 02 03 02 02 Total percentage with any leave 39 63 86 95 Paternity leave coverage: Percentage with unpaid leave 26 53 84 93 Percentage with paid leave 01 01 02 02 Total percentage with any leave 27 54 86 95 B. Part-time employees, medium-sized and large establishments Maternity leave coverage: Percentage with unpaid leave 19 36 42 54 Percentage with paid leave 01 01 00 00 Total percentage with any leave 20 37 42 54 Paternity leave coverage: Percentage with unpaid leave 14 32 42 54 Percentage with paid leave 00 01 00 00 Total percentage with any leave 14 33 42 54 C. Full-time employees, small establishments Maternity leave coverage: Percentage with unpaid leave 17 18 47 48 Percentage with paid leave 02 02 02 02 Percentage with any leave 19 20 49 50 Paternity leave coverage: Percentage with unpaid leave 08 08 47 48 Percentage with paid leave 00 01 02 02 Percentage with any leave 08 09 49 50 SOURCE: Bureau of Labor Statistics Employee Benefits Surveys (now called the National Compensation Surveys), various years, available from http://www.bls.gov/ebs/. The surveys define medium-sized and large establishments as those with 100 or more employees and small firms as those with fewer than 100 employees. Starting in 1994, figures are for family leave coverage rather than maternity or paternity leave. The passage of the FMLA in 1993 led to dramatic increases in both maternity and paternity leave coverage (see Tables 8-2 and 8-3). By 1997, nearly all full-time employees in medium and large firms had the right to a job-protected family leave, that is, leave for maternity, paternity, or to care for a newborn or newly adopted or placed child. However, because the FMLA covers only those who work in firms with 50 or more employees and who have worked 1,250 hours or more in the past 12 months, coverage rates among part-time employees and employees in small firms are much lower.
POLICIES TO SUPPORT WORKING FAMILIES 233 TABLE 8-3 Percentage of Public-Sector Employees with Family Leave Coverage, 1990 to 1998 1990 1992 1994 1998 A. Full-time employees Maternity leave coverage: Percentage with unpaid leave 51 59 93 95 Percentage with paid leave 01 01 04 04 Total percentage with any leave 52 60 97 99 Paternity leave coverage: Percentage with unpaid leave 33 44 93 95 Percentage with paid leave 01 01 04 04 Total percentage with any leave 34 45 97 99 B. Part-time employees Maternity leave coverage: Percentage with unpaid leave 28 32 62 56 Percentage with paid leave 01 01 01 01 Total percentage with any leave 29 33 63 57 Paternity leave coverage: Percentage with unpaid leave 18 24 62 56 Percentage with paid leave 01 01 01 01 Total percentage with any leave 19 25 63 57 SOURCE: Bureau of Labor Statistics Employee Benefits Surveys (now called the National Compensation Surveys), various years, available from http://www.bls.gov/ebs/. Starting in 1994, figures are for family leave coverage rather than maternity or paternity leave. Data for 1996 not available. Overall, the FMLA covers only about 60 percent of private-sector em- ployees, and only about 45 percent are both covered and eligible; about a quarter of those covered are not eligible due to short working hours or short job tenures (Commission on Family and Medical Leave, 1996; Cantor et al., 2001). Among all employees (public sector and private sector com- bined), the law covers about 75 percent, and just over 60 percent are both covered and eligible (Cantor et al., 2001). Coverage and eligibility rates vary a good deal by demographic characteristics, as shown in Table 8-4. Employees who are young (ages 18-24), have less than a high school educa- tion, or have low annual family income (less than $20,000) are much less likely to be covered and eligible than other employees. In addition to not providing universal coverage, the FMLA is limited in that it provides for only 12 weeks of leave, which is unpaid. In contrast, other countries provide longer periods of coverage and generally provide at least some wage replacement during the leave (see Table 8-5). Studies have found that women who have leave coverage are more likely to take a leave, and take longer leaves, but are also more likely to
234 WORKING FAMILIES AND GROWING KIDS TABLE 8-4 Family and Medical Leave Act Coverage by Demographic Characteristics, 2000 Percentage of Employees in Each Demographic Category Who Are: Covered and Covered Eligiblea Noncovered Gender* Male 74.9 62.3 25.1 Female 78.5 61.2 21.5 Age**++ 18-24 83.3 43.8 16.7 25-34 77.3 63.0 22.7 35-49 76.7 66.8 23.3 50-64 74.0 66.7 26.0 65 and over 58.7 42.8 41.3 Race/ethnicity**+ White non-Hispanic 73.5 59.7 26.5 Black non-Hispanic 93.3 71.8 06.7 Hispanic 80.2 66.2 19.8 Asian 92.0 73.4 -- All others 79.8 60.3 20.2 Marital status**++ Married/living with partner 74.3 63.8 25.7 Separated/divorced/widowed 79.3 64.3 20.7 Never married 82.6 54.1 17.4 Children under 18 in household None 78.2 60.9 21.8 One or more 74.8 63.2 25.2 Education**++ Less than high school 63.8 44.2 36.2 High school graduate 72.1 57.1 27.9 Some college 79.4 62.2 20.6 College graduate 77.1 65.3 22.9 Graduate school 88.0 73.8 12.0 Annual family income++ Less than $20,000 71.8 38.6 28.2 $20,000 to less than $30,000 78.8 64.5 21.2 $30,000 to less than $50,000 77.9 63.9 22.1 $50,000 to less than $75,000 79.7 70.2 20.3 $75,000 to less than $100,000 81.1 70.9 18.9 $100,000 or more 81.4 74.0 18.6
POLICIES TO SUPPORT WORKING FAMILIES 235 TABLE 8-4 Continued Percentage of Employees in Each Demographic Category Who Are: Covered and Covered Eligiblea Noncovered Compensation type**++ Salaried 78.7 70.8 21.3 Hourly 80.6 60.5 19.4 Other 52.1 37.7 47.9 aThe "Covered and Eligible" column is a subset of the "Covered" column. *Difference between covered and noncovered employees is statistically significant at p < 0.10. **Difference between covered and noncovered employees is statistically significant at p < 0.05. + Difference between covered and eligible employees and all other employees is statistically significant at p < 0.10. ++Difference between covered and eligible employees and all other employees is statistically significant at p < 0.05. Column percentages may not total to 100% due to rounding. SOURCE: Cantor et al. (2001:Table A2-3.4). return to work for their prebirth employer, than women who lack coverage (Glass and Riley, 1998; Han and Waldfogel, 2002; Hofferth, 1996; Joesch, 1997; Klerman et al., 1998a, 1998b; Ondrich et al., 1996, 1998; Ross, 1998; Waldfogel, 1999b). Studies have also found that the lack of paid leave is a barrier to women's taking leave, or taking as much leave as they feel they need (Cantor et al., 2001; Waldfogel, 2001c). Given that leave coverage is limited, unpaid, and of short duration, it is perhaps not surprising that mothers in America return to work much more quickly after birth than mothers in other comparable countries. A third of new mothers in the United States return to work within 3 months of giving birth, compared with only about 5 percent in Britain, Germany, and Swe- den; half of new mothers in the United States are back at work within 4 to 6 months, compared with 15 months in Sweden, over 24 months in Ger- many, and over 36 months in Britain (Gustafsson et al., 1996; Klerman et al., 1990, 1994, 1999; Smith and Bachu, 1999;). Data from the National Longitudinal Survey of Youth indicate that maternity leave coverage and usage vary somewhat by demographic char- acteristics of the family (Berger and Waldfogel, in press). Women with less than a high school education, for instance, are less likely than other new mothers to have the right to a job-protected maternity leave, less likely to
236 WORKING FAMILIES AND GROWING KIDS history; work sufficient if unpaid 3 weeks rate, 12 flat Countries for unpaid 6 rate (income-tested) rate months 3 rate flat wks benefit unpaid 18 pay, flat 3rd pay Comparable weeks, for 6 earnings 80% unemployment, child; earnings earnings earnings 55% earnings rate earnings earnings earnings @ @ 1 rate, 10 earnings earnings for @ prior flat for flat prior prior prior prior pay prior unemployment prior prior and rate of rate prior prior wks not, months months years Payment Unpaid 15 55% 90% if Unpaid 60% 90% 70% 70% Flat 80% Flat 12 100% 18 100% Unpaid 100% 2 80% 30% States United Months 2002. the in Total 2.8 12.0* 7.2 18.5 36.0 36.0 18.0 27.7 36.0 39.2 11.0 in months 3 12 Policies is to 3 is child leave Leave leave leave leave leave leave leave leave leave leave leave leave leave leave until child leave leave leave leave leave leave until leave parental family maternity parental maternity parental maternity maternity parental parental maternity maternity maternity maternity parental Leave parental childrearing parental leave parental childbirth-related of of weeks weeks weeks weeks weeks weeks weeks weeks weeks months weeks weeks weeks year years years years months months Type 12 17 35 18 13 28 1 18 26 Childrearing 52 2 18 16 2 16 Parental 14 3 5 6 period its (2001b). Childbirth-Related 8-5 extended Waldfogel States Kingdom try *Canada TABLE Coun United Canada United Denmark Finland Norway Sweden Austria France Germany Italy SOURCE:
POLICIES TO SUPPORT WORKING FAMILIES 237 A 29-year-old mother of two children describes the type of flexibility that would be needed to deal with family health problems (Harris and Lengyel, 2002:24): Holding a job was very hard for two reasons. It was my weight and the other was my son's disability. I would often need days off for my son's doctor's appointments or even weeks if he was hospitalized. I knew there wasn't an employer on earth that would be that flexible or understanding. So needless to say, I lost a lot of jobs. take any paid maternity leave, and more likely to return to work in the first six weeks after the birth. Never-married mothers, too, are more likely than other mothers to return in the first six weeks following birth. Surveys conducted post-FMLA have found that not all new parents who are covered by the FMLA take the 12 weeks to which they are entitled, while others take more than 12 weeks (presumably because their employer offers a more generous policy). The 2000 Westat Survey of Employees found that among those who took a maternity leave, the largest group (40 percent) took a leave that lasted between 6 and 12 weeks, but with substan- tial numbers taking a leave of less than 6 weeks (31 percent) or more than 12 weeks (29 percent) (Cantor et al., 2001). Medical Leave The FMLA also provides medical leave--leave that an employee can take because of her or his own serious illness or because of the serious illness of a family member, including a child. Coverage is limited to those meeting qualifying conditions, the leave is limited to only 12 weeks per year, and it is unpaid. Thus, although the FMLA has extended medical leave coverage to some workers who previously lacked coverage, it has not provided universal coverage, nor has it provided paid coverage. As of 1999, 47 percent of employees in the private sector did not have paid sick leave, and the rate of noncoverage was strongly associated with job charac- teristics, with 62 percent of blue-collar and service employees and 41 per- cent of clerical and sales employees lacking paid sick leave in contrast with only 19 percent of professional, technical, and related employees (Bureau of Labor Statistics, 2001). Previous research has found that parents who lack paid sick leave are less likely to stay home with a sick child than parents who have paid sick leave (Heymann, 2000).
238 WORKING FAMILIES AND GROWING KIDS TAX POLICIES In this section, we consider the major tax policies that help families with working parents to cover child care expenses. These include the child and dependent care tax credit and the Dependent Care Assistance Program. We also discuss the earned income tax credit, which although not a child care policy, is a major tax policy that supports low-income working fami- lies. We also briefly discuss the child tax credit, an important policy for families with children. Child and Dependent Care Tax Credit and Dependent Care Assistance Program Some relief for child care expenses of working parents through the tax code has been available in the United States since 1954. In fact, until recently, the dependent care tax credit, now known as the child and depen- dent care tax credit (CDCTC), and the employer-provided Dependent Care Assistance Program (DCAP) together constituted two of the largest federal programs for helping families with child care expenses.1 However, as discussed below, an important limitation of both these programs is that they have not reached many low-income families and have provided only limited support to the families they do reach. Currently, families are eligible for the CDCTC if they have earned income and maintain a household for a dependent under the age of 13 or for a spouse or other dependent (regardless of age) who is mentally or physically unable to care for himself or herself. In order to qualify, at least one spouse must be working and the other must be working or attending school full time. Also, the child care expenses must be work-related. As of 2003, the total amount of child care expenses that can be used for the CDCTC is $3,000 for one child or $6,000 for two or more children. Families can receive credit for up to 35 percent of their child care expenses up to these limits, with the percentage declining as family income increases to a base of 20 percent of child care expenses. The maximum credit is $1,050 for one child and $2,100 for two or more children (see Internal Revenue Service, 2002, for details). Thus, for families that spend more than $3,000 per child per year on child care, the maximum credit will cover a smaller share of their child care expenses. As indicated in Chapter 3, the 1Until the early 1990s, the federal government spent more on child care tax credits than on Head Start. Here we focus on the federal CDCTC and DCAP programs. There are also state- level programs in about half the states. Information about these is available from the Na- tional Women's Law Center (at www.nwlc.org).
POLICIES TO SUPPORT WORKING FAMILIES 239 average cost per week of care for a family that paid for care in 1999 was $76--or an average of $3,952 per year. This figure, however, represents the average, and some families spend more. The CDCTC is not refundable (and is thus not available to families whose incomes are so low that they do not pay taxes). Total CDCTC expenditures for 1999 were $2.675 billion distributed to 6.2 million house- holds, for an average benefit of about $430 per family (Campbell and Parisi, 2001). Largely because the tax credit is nonrefundable, it is regres- sive for low-income families (those earning less than $10,000 receive next to nothing). However, the tax credit is progressively distributed over most of the income distribution above the lowest income quintile (Gentry and Hagy, 1995). Data from 1997 indicate that about 10 percent of the credit went to families with adjusted gross annual incomes of less than $20,000, 42 percent went to families with incomes of between $20,000 and $50,000, and 48 percent went to families with incomes above $50,000 (U.S. House of Representatives, 2000). DCAP rules work somewhat differently. Basically, DCAP is a program that allows employees to choose to reduce their pretax income by a sum of up to $5,000 and use that money to pay for child care expenses. Thus, a DCAP benefits employees who will owe taxes by making a portion of their income (up to $5,000) tax exempt--but it does not offer any benefits to employees who will not owe taxes. Employers are not required to offer a DCAP program, but if they do, they must provide it as a part of an em- ployee benefit package available to all employees. Examples of qualified child care and dependent care expenses include payments to child care centers, work-related babysitting, domestic help, and nannies. Even grand- parents, uncles, aunts, and adult children qualify as child care providers if they are not also dependents of the tax filer. Federal expenditures on DCAP totaled nearly $1 billion in 2000 (Blau, 2001). The same child care dollar cannot be claimed for both the CDCTC and DCAP. Whether a family benefits more from one or the other depends on their income and marginal tax rates, but DCAP is used by fewer families because it is offered by a limited number of employers and because its benefit exceeds the CDCTC only for those in higher income brackets. As indicated above, neither program benefits families whose incomes are so low that they do not pay taxes. Earned Income Tax Credit The earned income tax credit (EITC) is a refundable tax credit for low-income working families. While not a child care policy, the EITC is a potentially important source of support for child care expenses for low- income families, particularly for those families whose incomes are so low
240 WORKING FAMILIES AND GROWING KIDS that they do not pay taxes and thus cannot benefit from the CDCTC or DCAP. The EITC has grown 300-fold since its inception in 1975. Changes in the program have occurred in three major periods. Through the Tax Re- form Act of 1986, the EITC increased by over 50 percent starting in 1987 and became indexed to inflation. With the Omnibus Budget Reconciliation Act (OBRA) of 1990, the value of the EITC increased by over 50 percent (phased in over three years, beginning in 1991), an additional credit was established for families with two or more children, and the EITC was no longer counted as income for most federal means-tested programs. With the Omnibus Budget Reconciliation Act of 1993, the EITC increased by over 50 percent (phased in over three years, starting in 1994), and a small EITC was established for taxpayers without children and age 25 or older. The total amount spent per year on the EITC was about $30 billion in 1999, of which about $26 billion was refundable (i.e., paid to individuals whose earnings were so low that they did not have to pay taxes). The average benefit was over $1,500 per family. A number of states have also implemented their own state EITCs, usually set as a percentage of the federal tax credit. Researchers are beginning to study the extent to which families know about and use the EITC, and how knowledge and usage vary by family characteristics; see, for instance, Ross Phillips (2002b), who found that low-income Hispanic parents are less likely than other low-income par- ents to know about and use the EITC. However, research on the extent to which the EITC is used to cover child care costs is limited. One study (Smeeding et al., 2000) asked families how they spent the money and found that few reported spending it on child care; however, in this study, some child care spending may have been reported under the category of education. Child Tax Credit The child tax credit (CTC) was until recently a nonrefundable tax credit available to families with children. In 2001, legislation was passed doubling the value of the CTC over the next 10 years (from a maximum of $500 per child to a maximum of $1,000 per child) and making the CTC partially refundable. Analysts from the Brookings Institution (Sawhill and Thomas, 2001) estimate that the new refundable benefits will provide about $540 per year on average to low-income families. The recent changes will mainly benefit families with annual incomes between $10,000 and $35,000 (and, to a lesser extent, large families with incomes above $35,000). Like the EITC, the CTC is not tied to child care expenditures, and little is known about how families use the credit.
POLICIES TO SUPPORT WORKING FAMILIES 241 EDUCATION PROGRAMS Public education is the biggest public support program for working parents, providing care for children ages 5 to 18 (and sometimes younger) for a substantial portion of the working day, even if not year-round. Nearly $300 billion was spent on public elementary and secondary education in 1999, serving about 47.2 million children (including kindergarten and some prekindergarten). Although there has been little research on the effect of public education on employment decisions, one study found large effects of the availability of kindergarten on the employment decisions of single moth- ers with 5-year-olds (Gelbach, 2002). It is important to note, however, that school schedules do not provide care during all the hours that parents work. Indeed, the mean number of hours that school-age children of employed parents are in care other than school is 21 hours per week (see Chapter 3, this volume). The education programs discussed here include public schools and pre- school and after-school education programs, which are supported through general funds and a number of other federal, state, and local funding streams. In the following sections, we consider the major education pro- grams that provide support for preschool or after-school care for working families, including full-day public kindergarten programs, Head Start, Title I preschool funding, the Individuals with Disabilities Education Act, and 21st Century Community Learning Centers. Full-Day Public Kindergarten Public kindergarten provides significant child care services for working families (85 percent of kindergarten is public) (Jamieson et al., 2001). Until recently, most kindergarten was provided for only part of the day (or part of the week). Since 1965, the number of children in kindergarten has fluctuated in a narrow range around 4 million; however, the rate of chil- dren in full-day (full school day hours) and full-week kindergarten has increased from about 1 in 10 to more than 1 in 2 (59 percent) in 1999 (Jamieson et al., 2001). Still, there are broad differences state by state (and sometimes from city to city) in the amount of full-day services offered (Mitchell, 2001). Eight states and the District of Columbia mandate full- day kindergarten programming in their public schools (Galley, 2001), but most leave it to local discretion. The primary impetus for the move to full-day kindergarten has been to improve educational outcomes for children, particularly those from disad- vantaged family backgrounds. Children who attend schools that have high proportions of children in poverty are much more likely to have full-day kindergarten programs, as are children who attend large schools (Love et
242 WORKING FAMILIES AND GROWING KIDS al., 1992). In addition, with more mothers working full time and more children entering kindergarten having already attended preschool full time, families have increasingly come to expect full-day kindergarten from their schools. There is some research to indicate that children who attended full- day (as opposed to part-day or part-week) programs had higher achieve- ment at the end of the year and in first grade (Fusaro, 1997; Gullo et al., 1986; Gullo and Clements, 1984; Koopmans, 1991). Public Prekindergarten As the number of children attending some form of preschool before kindergarten has increased,2 states have become increasingly involved in providing publicly funded prekindergarten programs. In the last decade alone, state funding for public prekindergarten has increased from $700 million (Adams and Sandfort, 1994) to about $1.9 billion in 1999 (Schulman et al., 1999; Mitchell et al., 1998; Education Week, 2001). A total of 39 states and the District of Columbia provide public prekindergarten for at least some 3- to 5-year-olds, up from 10 states in 1980; 21 states and the District of Columbia use state funding to serve additional children in Head Start (Education Week, 2001). However, state investments in prekindergarten vary considerably from state to state and there is a variation in the range in these services in terms of such things as who they serve, teacher training, class size, and curriculum In 1999, three- quarters of state funding of prekindergarten was concentrated in just 10 states, while 11 states spent no money on public prekindergarten or state Head Start (Schulman et al., 1999). There is only limited research to date on the characteristics of state-funded prekindergarten programs and their effects on child outcomes (Gilliam and Zigler, 2000; Ripple et al., 1999; Zigler and Styfco, 1993). Head Start and Early Head Start Started in 1965 as part of the war on poverty, Head Start's goals are broad: "to promote school readiness by enhancing the social and cognitive development of low-income children through the provision of health, edu- cational, social, and other services" (Head Start Act of 1998 U.S.C. 9801, et seq 1998). Head Start is federally funded with a 20 percent local match. 2Since 1965, the number of 3- and 4-year-old children attending some form of preschool has increased from 520,000 to 4,578,000 in 1999. The share of those services that is public has increased from less than 25 to 50 percent (Jamieson et al., 2001). Furthermore, a signifi- cant portion (a little less than half in 1999) of those publicly funded preschool services are provided on a full-time basis.
POLICIES TO SUPPORT WORKING FAMILIES 243 Some states invest their own funds above the required match. Head Start children receive free services, which include early education, social, and health and nutrition services. Targeted at low-income families with 3- and 4-year olds (and also expected to serve children with disabilities), Head Start served 905,000 children in fiscal year (FY) 2001 at a cost of over $6 billion. Of the children served in 2001, 54 percent were age 4 and 35 percent were age 3; 13 percent were children with disabilities (Administration for Children and Families, 2002). Head Start has increased more than fourfold since the mid-1980s, but it does not serve all eligible children. Currie (2001) found that Head Start served 50 percent of eligible 3- and 4-year-olds. This is a higher percentage of eligible children served than in the past, reflecting both program expansions and also reductions in the numbers of poor children eligible for the program. Starting in 1995, an Early Head Start initiative has served children under age 3. Early Head Start has expanded rapidly and now has projects in all 50 states (U.S. Department of Health and Human Services, 2002). Early Head Start served 62,000 infants and toddlers in FY 2002 (U.S. Department of Health and Human Services, 2002). However, this represents a small share of eligible children under the age of 3. Although Head Start is mostly a part-day, part-year program for poor families--only one-quarter of Head Start children receive full-day, full-year services (U.S. Department of Health and Human Services, 2002)--and one that involves a high level of parent involvement, many Head Start parents work. In 1995 (the most recent year for which data are available), 28 percent of Head Start parents were employed full time and 17 percent part time (Smith, 2000); these percentages are probably substantially higher now given the large increase in single mothers' employment in the 1990s. In 1995, almost one-third of enrollees used Head Start exclusively, and 79 percent used Head Start as their primary child care provider. For those who used multiple providers, most used relative care, primarily from grand- parents. Among users of nonrelative care, very few used day care centers or nursery schools (Smith, 2000). There have been several recent community- based initiatives to combine Head Start services with other child care ser- vices in the community, to better meet the needs of working families while providing high-quality services for children (see Schumacher et al., 2001b; Lombardi, 2003). Findings from the Early Head Start evaluation indicate that such collaborations can improve the quality of care (Administration for Children and Families, 2002). In accordance with its mandate, Head Start serves families with low incomes. The majority (77 percent) have incomes below $15,000 per year. Nearly equal proportions are black (34 percent), white (30 percent), and Hispanic (30 percent); relatively few were American Indian or Asian. Re- flecting the program's mandate to serve disabled children, 13 percent of
244 WORKING FAMILIES AND GROWING KIDS Head Start children had some form of disability (Administration for Chil- dren and Families, 2002). In recent years, there has been continued debate over the effects of Head Start on child outcomes. While there is a good deal of evidence to show that model early childhood intervention programs yield high and sustained benefits (see, for instance, Barnett, 1996; Ramey et al., 1999; Schweinhart et al., 1993; see also recent reviews by Currie, 2001; Karoly et al., 1998), the evidence on Head Start has been mixed. An early study (Westinghouse, 1969) found that gains in cognitive development faded after a few years in public schools, but subsequent research has found some positive long-term effects (Currie and Thomas, 1995, 1999, 2000; Garces et al., 2000). Currently, debate continues over the extent of positive effects associated with Head Start (U.S. General Accounting Office, 1997, 1998, 2000a,b). In large part, this debate stems from the lack of a randomized evaluation of Head Start. The U.S. Department of Health and Human Services recently launched such an evaluation--the National Head Start Impact Study, which is now under way (U.S. Department of Health and Human Services, 2002).3 Unlike Head Start, Early Head Start was implemented with a random assignment evaluation that included 17 programs in diverse communities around the country. Results indicate that at age 2 and at age 3 (when the intervention ended) the children in the program performed significantly better than control children on a wide range of measures of cognitive, language, and social-emotional development; in addition, the parents scored significantly higher than the control group on measures of the home envi- ronment, parenting behavior, and knowledge of infant and toddler devel- opment (Early Head Start Research Consortium, 2001; Love et al., 2002). While many of the effects were small (with effect sizes in the 10 to 20 percent range), the overall pattern of results, with significant differences across many types of outcomes for both children and parents, is indicative of the effectiveness of the program. Moreover, several subgroups demon- strated larger effects. For example, larger impacts were found among families in programs that used a mixed approach (a combination of home- based and center-based), especially programs that were fully implemented; families who enrolled during pregnancy; and families with a moderate number of risk factors (Administration for Children and Families, 2002.) 3The Department of Health and Human Services is also conducting a large-scale observa- tional study of Head Start, involving a nationally stratified random sample of 3,200 children and families in 40 Head Start programs. This study has found that Head Start narrows the gap between disadvantaged students and other children in key components of school readi- ness, with the largest gains for the children who had the lowest cognitive skills to start with (U.S. Department of Health and Human Services, 2001b).
POLICIES TO SUPPORT WORKING FAMILIES 245 Title I Preschool Title I was enacted under the Elementary and Secondary Education Act (1965) as a part of the war on poverty, to help schools meet the needs of children economically and educationally disadvantaged. Historically, most children served by Title I have been between the ages of 5 and 18, but recently Title I has expanded to include younger children. Much of the rise in services for preschoolers follows changes in eligibility regulations insti- tuted in 1994. (Prior to 1994, only schools with 75 percent or more of their students living in poverty could use their funds to improve the whole school. The 1994 reauthorization lowered the poverty eligibility threshold to 50 percent.) During the 1999-2000 school year, 17 percent of school districts that received Title I funding spent money on preschool services (U.S. General Accounting Office, 2000a,b). In that year, $407 million in Title I funds (out of more than $8.4 billion) went to an estimated 313,000 preschool children, or about 8 percent of children who will eventually enter kinder- garten. Title I funds a variety of services for preschoolers, including educa- tion, meals, medical, dental, and social services. In the largest districts, most children served are minorities (45 percent black, 39 percent Hispanic, 11 percent white, 3 percent Asian, and 1 percent American Indian). In smaller school districts, a larger percent of those served are white (35 percent) and fewer black (21 percent) (U.S. General Accounting Office, 2000b). Individuals with Disabilities Education Act In 1975, Congress determined that millions of American children with disabilities were not receiving an appropriate education: more than half of handicapped children did not receive services that would enable them to have full equality of educational opportunity. The Individuals with Disabilities Education Act of 1975 was enacted to remedy this by requiring that all students with disabilities receive free and appropriate public education. As the importance of early education for children's outcomes has be- come clearer, Congress has expanded the Act's mission to more fully serve younger groups of children with disabilities. The Education of the Handi- capped Act Amendments of 1986 established the Early Intervention Pro- gram for Infants and Toddlers with Disabilities (Part C). In 1998-1999, nearly 189,000 children from birth through age 2 were served under Part C. In addition, more than 573,000 3- through 5-year-olds with disabilities are served in preschool (U.S. Department of Education, 1999). Evaluations of the impacts of the Individuals with Disabilities Education Act on child and family functioning are now under way (see Rosman et al., 2002.)
246 WORKING FAMILIES AND GROWING KIDS 21st Century Community Learning Centers The federal government funds expanded academic enrichment for chil- dren attending low-performing schools through 21st Century Community Learning Centers (CCLC). The CCLC programs also provide youth devel- opment activities, drug and violence prevention programs, technology edu- cation, art, music and recreation, counseling, and other services. Thus, CCLC programs are a potentially important resource to meet the after- school needs of children with working parents. The CCLC program started in 1995 as a demonstration project with $750,000 appropriated by Congress. The program increased to $200 mil- lion in 1999 and to $1 billion in 2002 (U.S. Department of Education, 2002). In 2002, about 6,800 rural and inner-city public schools in 1,420 communities were participating as CCLCs. Eligible entities include local educational agencies, community-based organizations, other public or pri- vate entities, and consortia of two or more of such agencies, organizations, or entities. As of January 2002, the No Child Left Behind Act converted the CCLC authority to a state formula grant. Until then, the U.S. Department of Education made competitive awards directly to designated lead agencies. Under the reauthorized authority, funds will flow to states based on their share of Title I, Part A, funds. Current CCLC grantees will continue to be administered by and receive funding through the U.S. Department of Edu- cation. States will use their allocations to make competitive awards to eligible entities. States are required to make awards only to applicants that will primarily serve students who attend schools with concentrations of poor students. Also, states must give priority to projects that will target services to students who attend low-performing schools (U.S. Department of Education, 2002). While the CCLC permits funding to be used in support of after-school programs for adolescents, most funded programs provide service to chil- dren between the ages of 6 and 14, and attendance in after-school programs is typically highest for children under the age of 12 (Grossman et al., 2002; Vandell and Shumow, 1999). While participation in school-based or com- munity-based programs is typical for adolescents overall, low-income youth and those from traditionally defined minority groups have relative low rates of participation (see Chapter 6, this volume). State and Locally Funded School and Community-Based After-School Activity Programs School and community-based programs are an important source of after-school care for children of employed parents. Government funds
POLICIES TO SUPPORT WORKING FAMILIES 247 provide partial support for some youth activities (e.g., the Boys and Girls Clubs of America), but the federal government does not provide a stable source of funds to support enrichment activities for adolescents. A recent report (National Research Council and Institute of Medicine, 2002) de- scribed the funding for youth programs as fragmented and heavily depen- dent on grants from foundations and private agencies. Some after-school programs are funded mainly by parent fees or from other funding sources (such as charitable foundations), while others receive at least some support from state or local funding. Such programs have rarely had their own public policies or dedicated funding streams, but in recent years states are becoming increasingly involved in the funding of after-school programs. According to the National Conference of State Legislatures, in 2000-2001, at least 20 states passed legislation pertaining to after-school policies, in- cluding some that created new funding streams for after-school programs (National Conference of State Legislatures, 2002). After-school programs that are sponsored by individual states include California's After School Learning and Safe Neighborhoods Partnerships, which serve almost 30,000 children, and the Ohio School Age Child Care Project, which serves 2,500 children. Programs sponsored by local govern- ments, school districts, and private foundations include The After-School Corporation (TASC) in New York City (White et al., 2001), LA's Best Program in Los Angeles, San Diego's 6 to 6 Extended Day Program, and the Extended Services Initiative that has been implemented in 17 cities (Grossman et al., 2002). CHILD CARE FUNDING In this section, we consider the major child care funding streams that provide financial support to families or child care providers to offset the costs of child care used by families with working parents. These include the Child Care Development Fund, the Social Services Block Grant, Temporary Assistance for Needy Families child care funding and transfers, the Child and Adult Care Food Program, the Summer Food Program, and state child care programs. Most of these funding streams go to the states, where they are combined with state funds and then distributed, mostly through vouch- ers (with the exception of nutrition programs, which provide support to providers). Child Care Development Fund In 1996, the federal system of support for child care was overhauled with the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA). The act combined three separate federal
248 WORKING FAMILIES AND GROWING KIDS funding streams for child care for low-income families--the Child Care Development Block Grant, Title IV-A at-risk child care, and other Title IV- A funding, which included Aid to Families with Dependent Children, job opportunities and basic skills training (JOBS), and transitional child care-- into a new block grant, the Child Care and Development Fund (CCDF). Under the CCDF, low-income mothers on welfare are no longer guar- anteed child care assistance, and states have considerable latitude in defin- ing the rules for low-income families to get help with their child care costs. In order to be eligible, children must be under the age of 134 and must reside with a parent who is working or participating in education or train- ing or is in need of protective services. Under federal rules, children must be living in a family with income of at most 85 percent of the state median, but (as discussed below) states can set the income cutoff at a lower level, and in fact most do. Total expenditures in FY 2000 were $9 billion, including funds transferred from the Temporary Assistance to Needy Families (TANF) program (Mezey et al., 2002) (see below for a description of TANF and how funds can be transferred for child care). The CCDF consists of three separate funding streams, each with its own set of requirements. The discretionary fund, essentially the old Child Care Development Block Grant, is 100 percent federal with no state match required. Congress sets aside a portion of discretionary funds for specific uses: quality improvements ($171.5.6 million in 2003), infant and toddler care ($99.3 million in 2003), education, technical assistance, and research ($9.9 million in 2003) and child care resource and referral, school-age child care, and an information hotline ($19 million in 2003). The second stream, the mandatory fund, is a rough equivalent of pre-PRWORA federal Title IV-A spending in each state. These funds are also 100 percent federal with no state match requirement. The third stream, also a mandatory fund but requiring a state match, is based on historical pre-PRWORA expenditures. If states want matching funds, they must meet a maintenance of effort requirement and obligate all of their mandatory unmatched funds. The CCDF subsidizes child care services for eligible families mostly through certificates (vouchers), but also through contracts with providers. (States choose the extent to which they pay providers through vouchers or contracts, but parents must be given the option of receiving a voucher.) Parents can select any legal child care provider, as defined by each state. Providers may include family or friends, legal unlicensed family providers, and licensed family child care or child care centers. Providers must meet 4The regulations allow states to spend CCDF funds on children ages 13 to 19 if they have a physical or mental disability or are under court supervision.
POLICIES TO SUPPORT WORKING FAMILIES 249 minimum health and safety requirements as defined by the state. Relative providers may be exempted from some of the minimum regulations. As mentioned earlier, each state sets eligibility guidelines, including the income cutoff for families. Nationwide, most states have set income eligi- bility levels below the maximum level allowed under the CCDF (85 percent of the state median income). According to states' plans for FY 2002 and 2003, eight states established eligibility at the maximum level; the remain- der set their cutoffs at a lower level (Gish and Harper, 2002). Eligibility does not guarantee that a family is served. Despite having flexibility in defining eligibility, no state serves all its low-income children. According to a study by the U.S. Department of Health and Human Ser- vices (DHHS), only 10 to 15 percent of eligible families received any form of child care subsidy through the CCDF in 1999 (Administration for Chil- dren and Families, 1999, 2000), and the percentage varies considerably from state to state. About one-fifth of states are serving less than 10 percent of the children eligible for CCDF subsidies as defined by state eligibility criteria. Three-fifths are serving between 10 and 25 percent; and one-fifth are serving 25 percent or more (Administration for Children and Families, 1999).5 More recent studies have found similarly low percent- ages of children served (see, for instance, Collins et al., 2000, who found that states could serve 15 to 20 percent of eligible families, and Mezey et al., 2002, who estimated using DHHS data that states are serving 14 percent of eligible children.) There are various reasons why such a small share of eligible families is receiving child care subsidies through the CCDF. Lack of funding for eligible families is a fundamental issue. In some states, there are waiting lists for subsidies. In other states, however, there are either no waiting lists, or program officials report that they have sufficient funding to serve all the eligible families who have applied (U.S. General Accounting Office, 2001). This could be the result of a small applicant pool if states set eligibility levels very low. Or it may be that some families do not know that they are eligible, or they do not believe that the type of child care they would like to use would be covered. Many families report that they find the system of child care subsidies complex and difficult to navigate (Adams et al., 2002b). Lack of outreach and administrative barriers have also been noted (Adams et al., 2002b; Schumacher and Greenberg, 1999). Most importantly, the ability of a state to serve all eligible families who apply does not necessarily mean that all eligible families who want services are getting them, since 5A small number of states (for example, Illinois) have made a commitment to serve all eligible families; however, they accomplish this by limiting eligibility to only very low-income families (Waldfogel et al., 2001).
250 WORKING FAMILIES AND GROWING KIDS there is evidence that when new funds become available, more families apply (Adams and Rohacek, 2002). States set the copayment rates (the amount that parents will pay), on a sliding scale basis, which must take into account income and family size and may consider other factors as well. This fee can be waived for families below the federal poverty line and for children in protective services cases. CCDF regulations require that copayments must be affordable so that par- ents have access to a broad array of providers. In the past, DHHS has suggested that copayments should not exceed 10 percent of family income (Schumacher et al., 2001b); however, copayments differ widely around the country, and not all states have met this recommended level. States also have considerable flexibility in setting reimbursement rates, the amount that providers will be reimbursed for services. Under the CCDF, parents are to be given a full range of choices of care that is in accordance with state regulations, although state regulations (and prac- tices) vary considerably. As mentioned earlier, subsidized care can be provided by formal providers or by relatives and friends--that is, "informal care." If reimbursement rates are set too low, some providers will not accept subsidized children. So states must provide some evidence that their reimbursement rates meet the intent of federal law to provide choice to subsidized families and to provide equal access to care. States must conduct a market rate survey not more than two years prior to the effective date of the state plan and must establish payment rates for providers. DHHS guidance instructs states that a market rate set at the 75th percentile of the price distribution for care in that state will be considered adequate to meet the equal access requirement; however, many states do not meet that level. For instance, in a recent review of state plans, only 25 percent of the 56 states and territories included in the sample set their payment rates at or above the 75th percentile (Gish and Harper, 2002). Some areas set state- wide rates, but payment rates can vary by location, type of care, and age of children. The rate at which states set the rate ceiling affects how much of the provider's costs are covered by the state, which in turn affects how much subsidized child care is available to families (Adams and Snyder, 2003). Usage data on the CCDF is sparse and inconsistent from state to state. According to Administration for Children and Families 1999 data, the reason for care for almost 85 percent of children served was parent's em- ployment or a combination of employment and training or education. A small percentage of children received care either because their parent was in training or education (10 percent), in protective services (3 percent), or for some other reason (4 percent). By law, states are required to spend 70 percent of their mandated funds on families receiving, transitioning from, or at risk of becoming eligible for TANF assistance. According to CCDF plans for FY 2000 and 2001, more than half the states list TANF and
POLICIES TO SUPPORT WORKING FAMILIES 251 TANF-transitional families either first or second on their priority list of families who are eligible for receiving child care subsidies. Notwithstand- ing these priorities, a recent DHHS study that examined child care for low- income families in 25 communities nationwide found that, while states' funding policies favor TANF families over non-TANF families for receiving child care subsidies, children of non-TANF families represented the largest percentage of children receiving child care subsidies in most of the states that were examined (Abt Associates, 2000). The larger number of non- TANF families may reflect the fact that, as welfare caseloads have declined (due to both increased exits from welfare and decreased entries into wel- fare), child care subsidies have increasingly been taken up by families who are not on welfare. In FY 2000, approximately 1.7 million children were served by the CCDF (Child Care Bureau, 2000). Of these, just under a third (27 percent) were infants and toddlers, a quarter (26 percent) were older preschoolers, and over a third (36 percent) were school age. Most children were in state licensed or regulated care (74 percent), and the rest were in legally operat- ing but unregulated care (a little more than half of this latter group consists of relatives). Most CCDF-funded children are in centers (58 percent), and another 30 percent served in family child care. As discussed in Chapter 3, there is also some information on families receiving subsidies in the Survey of Income and Program Participation (SIPP). According to the latest published report on the SIPP data from the U.S. Census Bureau (2002), reporting data for 1997, only a small share of all families using child care receive subsidies or other government help toward the costs of the child care. Little is known at this point about the effects of child care subsidies on children and families. A small number of studies was conducted prior to welfare reform (see, for instance, Berger and Black, 1992; Meyers et al., 2002a; reviews by Blau, 2000, 2001). However, given the dramatic change in child care policies enacted under PRWORA, research that predates welfare reform is of limited use in understanding the likely effects of today's subsidy regime. Blau and Tekin (2001) provide some early evidence on the effects of child care subsidies post-PRWORA, finding that child care subsidies are associated with higher levels of employ- ment and school enrollment. Bainbridge et al. (in press) also provide some early evidence, examining the period 1991 to 1996, finding that subsidies directed at low-income working families had a substantial positive effect on boosting single mothers' employment over that period. As discussed in Chapter 3, an important characteristic of child care is its quality. Although the CCDF does not specify or control the quality of care that children receive, three of its provisions can affect the quality of services purchased with CCDF funds: health and safety protections, the quality set-aside, and the payment rates.
252 WORKING FAMILIES AND GROWING KIDS First, with regard to health and safety, states must document that they have established health and safety requirements that providers serving CCDF-funded children must meet (although most relatives would be ex- empt from these). At a minimum, these health and safety requirements must address the prevention and control of infectious diseases, immuniza- tions, building safety, and training for providers in the area of health and safety. Second, according to the federal law, 4 percent of CCDF funds must be set aside for quality improvements. According to a U.S. General Ac- counting Office (2002) study (based on case studies of 5 states plus a survey of states to which 42 states replied), the majority of states reported spend- ing more than the minimally required 4 percent. Of the funds that states spent on quality improvements, most (61 percent) came from CCDF and TANF, with about a third (29 percent) coming from state funds. In the 34 states that tracked the type of providers receiving the quality improvement funding, two-thirds of the expenditures went to child care centers, and less than a third to family child care or after-school programs. States spent their quality funds on a variety of improvements, including child care resource and referral services (20 percent of states), enhanced licensing inspections (14 percent), meeting state standards (13 percent), caregiver compensation (12 percent), off-site caregiver training (11 percent), incentives for accredi- tation (8 percent), on-site caregiver training (2 percent), and other activities (12 percent) (states could report spending in more than one area).6 Third, as discussed above, CCDF gives states a great deal of latitude to set payment rates. To the extent that higher quality care is more costly to provide, states can, through their payment rates for providers, affect the quality mix that is purchased with their CCDF funds. At the same time, to the extent that high copayments would discourage families from participat- ing in the program or from using more expensive forms of care that might be of higher quality, states can also affect the quality mix through the rules they set for copayments. (Copayments may be set as a percentage of family income, in which case high copayments may discourage families from par- ticipating; copayments may also be set as a percentage of the cost of care, in which case high copayments may also discourage parents from using higher priced care). 6 The extent to which these quality improvements are improving child outcomes is unclear. While many of the state initiatives are targeting reforms that have been found in other re- search to be associated with better child outcomes, the GAO study concluded that few of the state studies had sufficient evidence to draw conclusions about child outcomes. Therefore, the study recommended that DHHS include selected state quality improvement initiatives as part of a larger impact evaluation of state child care subsidy approaches (U.S. General Ac- counting Office, 2002).
POLICIES TO SUPPORT WORKING FAMILIES 253 Social Services Block Grant Child Care Starting with Title XX in 1974 and continuing as the Social Services Block Grant of 1981 (SSBG), states have received federal funds to assist them in delivering social services for adults and children. States have discretion in how they use these funds within a broad set of guidelines, and many states have used a substantial portion to fund child care for low- income families. In 1999, the total SSBG block grant was about $3 billion, of which at least $400 million (13 percent) was spent on child care services. The $3 billion total includes a TANF transfer of $1.17 billion (discussed below). Whether the SSBG should be considered a growing or diminishing source of child care funding is unclear. Data from 23 states in 1990 indicated that they spent 16 percent of their SSBG funds on child care (U.S. House of Representatives, Committee on Ways and Means, 1994). At the time, child care advocates raised concerns that states significantly reduced their SSBG funds for child care when federal funds under the Family Support Act and the Child Care and Development Block Grant began flowing. Certainly, current child care spending under the SSBG is not nearly as high as it was during the mid- to late 1970s, when over $700 million was spent annually on child care (U.S. Department of Health, Education, and Welfare, 1978). However, recent SSBG reports indicate that child care spending may be on the rise again since at least 1998, thanks largely to TANF transfers (U.S. Department of Health and Human Services, Administration for Children and Families, 1999, 2000a, 2001a). Data on SSBG usage has always been sparse. No data are available on who uses services--their income, work status, race/ethnicity, or education. Neither is there research on the impact of SSBG expenditures on children and families. TANF Child Care and Transfers TANF has grown to become a significant source of child care funding. In 1999, $604 million of TANF money was spent directly on child care for families receiving welfare payments. These funds were subjected to all the TANF restrictions. Under law, portions of TANF money can also be transferred to the CCDF or the SSBG, transfers that then have to be spent under CCDF or SSBG rules. In 1999, $2.43 billion was transferred from TANF to the CCDF, and $1.17 billion was transferred from TANF to the SSBG. In 2001, $1.88 billion was transferred to the CCDF, making TANF a very large source of child care funding. TANF transfers accounted for 27 percent of all CCDF child care subsidies in 1999. For the SSBG funds, it is not clear
254 WORKING FAMILIES AND GROWING KIDS what percentage of the transfer went to child care services, but dollars transferred from TANF accounted for 38 percent of overall expenditures under SSBG in 1999. TANF funding of child care is important (and perhaps a concern) for those who receive services in several important ways. First, if the child care is provided directly under TANF, the CCDF rules for regulation and quality do not necessarily apply. Second, even for transfers to the CCDF or the SSBG, availability of care may be jeopardized down the road as demands on TANF funding grow. Furthermore, uncertainty about future federal TANF funding levels can make it difficult to conduct long-run state child care policy planning (Schumacher et al., 2001a). Child and Adult Care Food Program The Child and Adult Care Food Program (CACFP) is an open-ended, federal subsidy for meals, snacks, and nutrition education in licensed child care centers (including Head Start), family and group day care homes. The program resembles the school meals program in being administered by the U.S. Department of Agriculture (USDA) and reimbursing meals according to a flat fee. In 1999, the program served about 2 million children at a cost of $1.4 billion (U.S. Department of Agriculture, 2002a). The primary target of the program is children whose income falls below 185 percent of the federal poverty level. However, subsidies for meals served in qualifying family home day care settings are not conditioned by the income levels of the children's families. In contrast, full subsides are paid in child care centers only for meals served to children from families with incomes 130 percent or less of the poverty level, at reduced prices for those at 130-185 percent of the poverty level, and with the smallest subsi- dies paid for meals served to children whose families earn over 185 percent of the federal poverty line. Comprehensive information about CACFP comes from a national sample survey conducted in 1995 (Glantz et al., 1997). Of the average daily 2.3 million children served meals in FY 1995, two-fifths were in family home day care, and the remainder were enrolled in Head Start or day care centers. Summer Food Service Program The Summer Food Service Program (SFSP) was created to ensure that children in lower-income areas could continue to receive nutritious meals when they do not have access to the National School Lunch or School Breakfast program. About 2 million children receive the free meals pro- vided during the summer months, at a cost of $238 million in 1999. Ap-
POLICIES TO SUPPORT WORKING FAMILIES 255 proximately 14 percent of the children who receive reduced-price or free school lunches participate in the SFSP (U.S. Department of Agriculture, 2002b). Program sponsors receive payments for serving meals and snacks to children and teenagers, 18 years and younger, at approved sites in low- income areas. Schools, public agencies, and private nonprofit organiza- tions may apply to sponsor the program. Potential sponsors must demon- strate that meal sites will meet either geographic or enrollment criteria. A site is geographically eligible if it is located in an area in which 50 percent of the children qualify for a free or reduced-price school meal. A site is enrollment eligible if 50 percent of the children enrolled can be documented to qualify for a free or reduced-price school meal (family income up to 185 percent of poverty). In addition, anyone attending a school program for people with disabilities, regardless of age, may also participate. Meals are served free to anyone at a site. State Funding for Child Care Programs As noted above, the various federal child care programs give states a great deal of latitude on how they spend federal child care funds. An- other important source of variation across states is state funding for child care programs. States choose not only how to spend federal dol- lars but also how much to invest in state dollars. The 1990s saw an expansion of state funding initiatives in this area (Meyers et al., 2002a; Adams and Rohacek, 2002). However, with contracting state budgets, these expansions may not continue. Indeed, several states are currently projecting that they will cut back their funding in this area (see, for example, Goodnough, 2003). STATE CHILD CARE REGULATIONS AND MONITORING Another important aspect of child care policy involves child care regu- latory and monitoring activity by the states. States play a potentially im- portant role in the child care arena by establishing regulations for child care settings and by monitoring the extent to which child care providers comply with those regulations. States vary a good deal in which settings are subject to state regulation and in the tightness of those regulations; states also vary a good deal in the intensity with which they monitor providers' compliance with those regulations (for useful discussions, see Blau, 2001; Gormley, 1999). It is also important to note that the goal of state licensing is to ensure basic health and safety protections, not to ensure quality. States vary in the level of these protections. For example, only two states regulate
256 WORKING FAMILIES AND GROWING KIDS all the provider types, and many states monitor programs once a year or less (U.S. General Accounting Office, 2000a). It is not clear how much impact this state regulatory and monitoring activity has on the child care market. There is some evidence that regula- tions do have an effect on the type and quality of child care that is offered. For instance, a recent study by the Early Child Care Research Network of the National Institute of Child Health and Human Development (2002c) found that regulations regarding staff training and staff-child ratios affect process quality in child care settings and ultimately child outcomes. How- ever, there is also evidence that, in many settings, child care regulations are not binding (Blau, 2001). There is also widespread agreement that any consideration of the role of regulations must take monitoring activity into account, since regulations that are on the books but not enforced will have little impact (Gormley, 1995). Recently, there has been increased interest in more nuanced monitor- ing systems that would allow states to differentiate between providers that just meet minimal standards and those that meet higher quality stan- dards (Blau, 2001; Gormley, 2000). At the same time, however, concerns have been raised that tighter regulations will raise costs at the same time as quality and thus may price lower income families out of the child care market (see, for instance, Currie and Hotz, 2001) unless additional funds are forthcoming. MILITARY CHILD CARE In 1989 the Military Child Care Act was passed to address issues of quality, program oversight, affordability, and availability of child care for families in the military. The act increased the amount of appropriated funds and increased inspections, regulations, training, and pay for child care staff. In the ensuing decade, the U.S. Department of Defense put energy, skill, and resources into creating a high-quality employer-supported child care system. A comprehensive system of child care options was established, with key components to ensure accountability and oversight and mandates for accreditation, training and professional development, wage enhancements for staff, and parent involvement. The Department of Defense is recognized today as a model for the nation in terms of its high-quality child care system. It has established over 800 child development centers for military personnel in over 300 locations worldwide, and 98 percent of these programs are accredited. Each day, the Department of Defense cares for over 200,000 children in centers, family child care homes, and school-age programs.
POLICIES TO SUPPORT WORKING FAMILIES 257 EUROPEAN POLICIES TO SUPPORT WORKING FAMILIES While the policies from other countries may not be fully replicable in the United States, Europe's family and work policies provide a useful com- parison of the extent to which working families are supported. Public policies in the United States provide less public support for the care of children than in other countries, relying more heavily on family members, nonpublic child care centers, and employers and less heavily on direct public provision. In the area of family leave, for instance, the United States mandates that certain employers provide leave, rather than publicly fund- ing leave, as is the case in most European countries. In the area of child care policy, the United States supports child care mainly through subsidies to parents or private market providers, rather than through public or publicly subsidized child care programs. A second point of difference is that programs in the United States tend to be more narrowly targeted to low-income families, rather than provided universally to all families with children. In the area of child care policy, for instance, the United States mainly targets assistance to the very lowest income families, but it does not reach all such families with its programs. Relying on family members, private child care providers, and employers may have advantages, particularly with regard to parental choice and flex- ibility, but it also creates challenges in terms of ensuring equality of access and the quality of supports provided. Targeting resources to the lowest income families also has advantages in terms of efficiency, but it creates challenges in terms of ensuring the quality of programs in which children are placed and ensuring equal access to high-quality care. These challenges are particularly acute in the areas of family leave and child care. In the area of family leave, because small firms are generally exempted from leave legislation and because the laws typically provide only unpaid leave, the United States has a system in which some new mothers lack the right to a job-protected maternity leave, and many face the loss of a substantial portion of their income if they take a leave. Those who lack leave tend to be the most disadvantaged; less educated and low- income workers are significantly less likely to be covered by the FMLA than other workers (see Table 8-4). Perhaps as a result of the limited leaves available, new mothers in the United States return to work much more quickly than do new mothers in other countries. Such early returns may pose risks for child health and development, particularly if the mother works full time (Ruhm, 2000a; Waldfogel et al., 2002). Accordingly, many analysts in the United States have called for expanding maternity (and paternity) leave coverage, extending the duration of leave allowed, and making some provision for income replacement during leave (see, for
258 WORKING FAMILIES AND GROWING KIDS TABLE 8-6 Share of Children in Publicly Supported Child Care, Selected Peer Countries Share of Children in Publicly Share of Costs Covered Supported Care by Government Children Ages Children Ages Children Ages Children Ages Country 0-2 3-6 0-2 3-6 United States 05 54 25-30 25-30 CaN/Ada 05 53 N/A N/A Denmark 48 82 70-80 70-80 Finland 21 53 85 85 Norway 20 63 68 68 Sweden 33 72 82-87 82-87 Austria 03 80 N/A N/A France 23 99 72-77 100 Germany 02 78 N/A N/A Italy 06 91 N/A N/A United Kingdom 02 60 N/A N/A N/A= not available. SOURCE: Waldfogel (2001a). example, Kamerman, 2000; Waldfogel, 2001b). Analysts have also called for expanded family and medical leave (see, for instance, Asher and Lenhoff, 2001; Heymann, 2000) and for more opportunities for mothers of young children to return to work part time (National Research Council and Institute of Medicine, 2000; Brooks-Gunn, Han, and Waldfogel, 2002). In the area of child care, the United States has a system in which parents must pay directly for most types of care and in which low-income families are not guaranteed assistance with the cost of care. Many European coun- tries, in contrast, guarantee a public or publicly subsidized child care place for any child whose parent wishes one from the age of 3, and several countries are now lowering the age at which child care is guaranteed to 1 or 2. Table 8-6 presents rates of enrollment in publicly supported child care in various countries. Even in Britain, which has had a strong tradition of exclusive maternal care until school entry at age 5, public opinion has shifted in favor of preschool experience beginning at age 3, and the govern- ment has made a commitment to guarantee at least a part-time child care place to each 3- and 4-year-old whose parent wishes them to have one (Hills and Waldfogel, 2002). It is likely that these European countries provide more equitable care than the United States, where affluent families can purchase high-quality care for their children, some low-income families may be fortunate enough to gain a subsidized place in high-quality care, but
POLICIES TO SUPPORT WORKING FAMILIES 259 children from other low-income and middle-income families tend to experi- ence lower quality care (Phillips et al., 1994). SUMMARY The public sector has responded to the challenges faced by working families in caring for their children by providing greater resources for them. Many important new public programs for children and adolescents have developed in the past 25 to 30 years in response to the increasing movement of mothers into the labor force. There has also been an expansion of social welfare programs to cover such services as early childhood education and medical care for low-income children. There is a growing recognition in the United States that early childhood care and education can confer advantages in terms of children's school readiness. Several groups have recently called for a move to universal prekindergarten (see, for instance, Committee on Economic Development, 2002). However, many of these programs are still not specifically designed to enhance the cognitive, social, and behavioral development of children. Those that do are not available to all children and adolescents. Furthermore, only 45 percent of parents working in the private sector have guaranteed unpaid parental leave through the FMLA. Less than 5 percent have access to paid parental leave. Many parents do not have the right to the more than 12 weeks of leave mandated by the FMLA. Fundamentally, policies and programs for working families and their children often focus on only one half of this equation--the employment of the parent or the well being of the child--without taking into consideration the simultaneous and interactive needs of both. There is a need for both policies and research that consider the needs of both parents and the chil- dren and adolescents in working families. Our review of the evidence on the role of public policies suggests two other priorities in the area of child care policy. One is the importance of guaranteeing funding for subsidies for low-income families with employed parents. The other is the need for efforts to raise the quality of child care that children in the United States experience, although analysts continue to debate the most effective means of doing so. In terms of research priorities, it is striking that for the majority of policies discussed in this chapter, very limited data are available on the effects on children. While this lack of data is being addressed in some areas with new evaluations under way, there clearly is a need for further data that would allow researchers to learn how the U.S. choice of public policies to support working families is affecting outcomes for their children.